Optimism and Attributional Retraining

8/3/2019 Optimism and Attributional Retraining

http://slidepdf.com/reader/full/optimism-and-attributional-retraining 1/22

Optimism and Attributional Retraining: Longitudinal

Effects on Academic Achievement, Test Anxiety, and

Voluntary Course Withdrawal in College Students’

JOELLEc.RUT HIG,2 RAYMOND P. PERRY,

NATHAN .HALL,AND STEVEN HLADKYJ

Universi@of Manitoba

Winnipeg, Manitoba, Canada

A longitudinal study exam ined how op timism an d attributional retraining (A R) influenced

256 first-year college stud ents’ test anxiety, cum ulativ e academic achievement, and course

persistence in college over an academic year. Students’ optimism was assessed a t the start

of the academic year and they were assigned to either an AR or no-AR (control) condition.

Measures of students’ test anxiety, cum ulativ e grad e point average, a nd voluntary course

withdrawal were obtained at the end of the academic year. Results suggest that although

high optimism was an academic risk factor am ong students who did not receive AR , high-

optimism students who did receive the AR cognitive intervention benefited from its

effects to a greater extent than did low-optimism students.

The initial transition to college introduces a multitude of challenges into the

lives of students that continues throughout their academic development. Specific

institutional and situational factors com bine w ith individual differences to influ-

ence how students adapt to the college experience, w hich, directly and indirectly,affects their long-term ac adem ic success. Two of these-namely, optim ism,

which is an individual-difference variable, and attributional retraining, a cogni-

tive intervention-each have been found to affect college studen ts in making the

transition to university and beyond (Perry, Hechter, Menec, & Weinberg, 1993;

Peterson, 2000). The current longitudinal study examines optimism and attribu-

tional retraining in terms of how, separately and in combination, they influence

students’ test anxiety, cumulative grade point average (GPA), and persistence in

college over the course of an academic year.

‘This study was supported by a research grant (410-99-0435) to the second author from the

Social Sciences and H umanities R esearch Council. Parts of the research in this study were presented

at the annual meeting of the American Educational Research Association in Seattle, Washington,

April 2001.

2Correspondence concerning this article should be addressed to Joelle C. Ruthig, Department

of Psychology, University of Manitoba, Winnipeg, Manitoba R3T 2N2, Canada. E-mail: umkobyla@

cc.umanitoba.ca

709

Journal of Applied Social Psychology,2004,34, 4, pp. 709-730.

Copyright0 004 by V. H. Winston & Son, Inc. All rights reserved.



710 RUTHIG ET AL.

Dispositional Optimism

A s a concept, optimism h as its origins in a form of folk wisdom that concerns

whether people expect good (or conversely bad) things to happen to them (C arver

& Scheier, 2000). This difference between individuals, representing the power o f

possibi li ty , of ten can shape the m eaning of thei r l ives (Dom ino & Conway,

2000), leading S cheier and Carver (1 98 7) to describe this “general expectation

that good things w ill happen” (p. 171 ) as dispositional optimism. Dispositional

optimism relates to positive expectations regarding possible outcomes such that

optimistic individua ls are confiden t that they w ill attain their goa ls successfully

(Carver, R eynolds, & Scheier, 199 4; Peterson, 2000; Scheier & Carver, 1993).

Since Scheier and Carver’s ( 1 985) early work with dispositional optimism, it has

gained populari ty a s a personali ty construc t am ong psychologists who h ave

studied its impact in various con texts, ranging from aca dem ic perform ance to

health and aging (e.g., Helgeson, 1999; Lai & Wan, 1996; Lennings, 2000).

Being opt imis t ic can have many posi t ive consequences for individuals

through enhancing physical and psych ological well-being, decreasing traumatic

distress, and increa sing motivation and m arital satisfaction (Do ugall, Hyman,

Hayw ard, McFeeley, & Baum, 200 1 ; Scheier & Carver, 1987; Taylor & Brown,

1988). Dispositional optimism is also associated with active coping, effectiveproblem solving, resilience in the presence of stressful life events, and, in the col-

lege setting, it has been associated with academ ic success (Aspinw all, Richter, &

Hoffmann, 2000; Floyd, 1997; K ao & Tienda, 1995; Peterson, 2000; Scheier &

Carver, 1987). Optimists also tend to attribute their problems to temporary, spe-

cific, external causes, rather than to stable, global, internal causes (an insidious

pattern characterist ic of helplessness in fai lure si tuations; Gil lham, Shatte,

Reivich, & Seligman, 2000).

Considering that this personality construct has a solid link to expectations,

motivation, and performance, it is no surprise that individual differences in dis-positional optimism are believed to be important for college students’ academic

development. T he college experience is full of adversities an d stressful situations

for students to deal with, and wh ile som e students tend to give up w hen faced

with acade m ic hardsh ips, optim istic studen ts m ay be m ore persistent because

they believe they can attain their goals eve n w hen difficulties arise (Peterson,

2000).

Costs o f Optimism

Although i t appears that being optimistic is positive in many ways, some

researche rs have found that optim ism can a lso be co stly if it is too unrealistic

(Peterson, 2000). For instance, Weinstein (1 98 9) foun d that optimistic individ-

uals tended to underestimate the likelihood of falling ill or having aversive events



OPTIMISM AND ATTRIBUTIONAL RETRAINING 711

occur in their lives. As a result, these overly optimistic individuals consequen tly

may fail to take proper health precautions and may engage in more risky behav-

iors (e.g., sm oking, unprotected sex), as compared to less optimistic individuals.

This d ifference in behavior appears to be a result o f the attitude o f overly opti-

mistic individuals that “bad things w on’t happen to m e” o r essentially viewing

themselves as invincible (Fromme, Katz, & Rivet, 1997). Thus, w hile it may be

beneficial not to worry needlessly abou t potential disaster every mom ent o f one’s

life, it may also be detrimental no t to be prepared when som ething negative does

happen.

Un realistic optim ism may occur for several different reasons. Higgins, St.

Am and, and Poole (1 997 ) cite cognitive and motivational m echanisms as deter-

minants of unrealistic optimism , two examp les of cognitive mechanisms being

egocentrism (i.e., believing oneself to be less vulnerable than other people) and

representativeness or the salience of the prototype (Weinstein, 1980). The moti-

vational mechanism involved in unrealistic optimism is self-deception, whereby

individu als disregard or distort the possibili ty o f bad things occu rring in an

attempt to reduce anxiety. Thus, unrealistic optimism m ay result from both cog-

nitive and motivational mechanisms and may affect health, social, criminal, and

academic outcomes (H iggins et al., 1997).

O f interest to us is em pirica l evid enc e indicating th at unrealistic o ptimismoccurs within the achievement context. Although com mon sense would sugg est

that optimistic students would strive harder to reach their goals than would less

optimistic students, and thereby d o better, it is also possible that overly optimistic

students are more likely to disengage from acad emic work because they set unre-

alistically high goa ls or inflate their perceived acad em ic abilities. Ro bins and

Beer (2001), for examp le, found that so m e overly op timistic students withdraw

psychologically from the academ ic contex t (i.e., view grades a s progressively

less important) in order to protect themselves from the negative effects of failure,

but they also may need to d isengage because they initially set unrealistically highgoals. Further, these unrealistically high goals m ay be based on unco ntrollable

causa l attributions (e.g., luck, ability), resulting in redu ced feeling s of control

over academ ic outcom es. Providing attr ibutional retraining in these circum-

stances may allow o verly optimistic students to transform their maladaptive

casual attributions into ad aptive causal attributions. Specifically, by replacing

uncontrollable, stable causal attributions with controllable, unstable attributions

(e.g., effort, strategy), the disengaging tend encies of unrealistic o ptimists may be

minimized and their feelings o f control restored. M ore will be said about the pos-

sible benefits of attributional retraining in a later section of this paper.Another drawback to un realistic optimism is the possibility that students may

expect only good things to happen and thus fa i l to form ulate a backup plan

when less desirable outcom es occur. In addition, highly positive illusions have

been found to decrease ego involvement, self-esteem, and well-being during the



712 RUTHIG ET AL.

academ ic year, though they m ay be beneficial to students in the short term a s an

ego-protection mechanism (Colvin & Block, 1994; Paulus, 1998). Robins and

Beer (2001), for example, show ed that a self-enhancement bias did not promote

superior academic performance amon g co llege students, and optimistic students

did not receive higher grades than did less optimistic students. Thus, they argue

that many of the positive effects of optim ism discove red by researchers “may

reflect a g eneral tendency to b olster self-esteem by denying inform ation that

threatens self-worth, . . . reflecting defensive denial rather than actual psycho-

logical adjustment” (p . 340).

Although there is empirical support for both perspectives on optimism, the

existing research has not yet fully resolved under w hat circumstances optimism isa benefit or a liability. Thus, the ex tent to which positive illusions are functional

rem ains open to debate. Adapting W einer’s (198 6, 1995) attribution theory to

exam ine optimism in terms o f causal attributions may provide som e clarification

of the issue. Specifically, it is possible that optimism is beneficial because high-

optimism students may attribute posit ive and negative academ ic outcomes to

controllable, unstable causes, s uch as effort or strategy. These attributions w ill

enhance achiev em ent striving because they enab le those co nditions producing

success to be repeated and those producing failure to be changed. Optimism m ay

be harmful, however, if high-op timism stude nts instead attribute positive andnegative outcomes to uncontrollable, stable causes, such as ability. In the latter

case, it may be a benefit to attribute success to high ability, which is an uncon-

trollable, stable cause, but in the event of failure, which is much m ore com mon in

the first year of college, a similar uncontrollable, stable attribution (low ability)

can be very detrimental to further achievem ent striving. Thus, previously dispar-

ate research findings regarding optimism m ay b e d ue to differences in causal

attributions for positive and negative outcomes.

It is important to no te that the current study involves a unique group of partic-

ipants; namely, first-year college students, whose experiences with the realities o funivers i ty l i fe can be extremely l imi ted. These low-control environments

(Thompson, S obolew -Shu bin, Galbraith, Schwankovsky, & Cruzen, 1993) are

comp rised of a variety of novel experiences, depending on the student, such as

(a) an emph asis on success-failure , (b) heighten ed acade m ic compet i t ion,

(c) increased pressure to excel, (d) more frequent academ ic failure, (e) unfamiliar

academic tasks, ( f ) critical career choices, and (8) new social networks. Thus, the

transition from high school into college can be extremely stressful on students

when considering their na’ivete in terms of what to expect academ ically. In this

regard, Perry, Hladkyj, Pekrun, an d Pelletier (2001)proposed a paradox o f failurein which bright enthusiastic high schoo l studen ts often are unable to adjust to the

increased dem and for self-initiative and autonomy. Thus, it is possible that high

optimism can be a risk factor for this population because they have little previous

experience on which to b ase their posit ive academic expectations. If so, high




optimism m ay be a g reater risk factor among students with no previous college

experience compared to second- or third-year college students, who m ay be high

in optimism, but w ho also have more realistic ex pectations, having experienced

previous academic successes and failures. Further, if highly op timistic students

attribute their academ ic outcom es to uncontrollable, stable causes (e.g., ability),

they run the r isk o f exp eriencing severe disappointment and ap athy once they

begin to encou nter failure with som e regularity. Consequently, by altering the

maladaptive (uncontrollable) causal attributions of optimistic first-year college

students to adaptive (controllable) causal attributions, the ben efits of optimism

may be maximized and the costs nullified, resulting in fewer subsequent failure

experiences.

Attributional Retraining

Attributional retraining, based o n W einer’s (1986, 1995) theory o f achieve-

ment motivation and emotion, is designed to enhance academic motivation and

achieve m ent striving. Acc ording to W einer’s theory, individu als continu ously

seek to explain outcom es a nd events in their daily lives. W ithin the college set-

ting, these explanations or causal beliefs about academic performance influence

subsequent motivation and ach ievement among students (Perry et al ., 1993).Poo r perform ance on a test attributed to lack o f ability, an un controllable and

stable cause, will lead to lower m otivation to perform in the future. In contrast, if

poor performance is attributed to low effort, a controllable an d unstable cause,

motivation to invest more effort into future performance is enhanced. Thus, attri-

butional retraining focuses on inducing effort attributions for failure and related

unstable, controllable causes, thereby increasing students’ perception of control

over their academ ic performance (Menec et al., 1994).

Research by Perry an d his colleagues (M enec et al., 1994; Perry et al., 1993;

Perry & Struthers, 1994; Struthers & Perry, 199 6) has s how n that attributionalretraining leads to enhanced motivation, increased perceived control, and better

academic performance amon g college students in both laboratory and actual

classroom settings. Although som e research suggests that attributional retraining

should benefit all types o f students, Wilson and Linville (1982 , 1985) found indi-

vidual differences am on g students in terms of the effectiveness of this psycho-

therapeutic intervention. Sub seque nt research h as revealed that at tributional

retraining is most effective when ap plied to at-risk studen ts (Perry & Penner,

1990). M enec et al., for example, found that co llege students who had low per-

ceived control (external locus o f control) over their academ ic performance per-formed better following attributional retraining, while high-control ( internal

locus of contro l) studen ts did not. Similarly, Perry and Struthers found that

college students with low perceived success regarding academ ic performance

benefited from attributional retraining, wh ereas their high perce ived-suc cess



714 RUTHIG ET AL.

counterparts did not. That is, for low-success students, those who received attri-

butional retraining (AR ) outperformed a no-AR control g rou p by more than on e

letter grade in a full-year psychology course. Th us, it appea rs tha t wh ile AR can

have positive effects on subsequent academ ic performance of college students in

general, those students wh o are at risk academ ically m ay benefit the most from

this cognitive intervention.

The current study examines whether optimism is adaptive in an academic set-

ting when com bined with a psychotherapeutic intervention involving AR . Based

on past findings that AR is beneficial for certain at-risk students (Menec et al.,

1994; Perry & Struthers, 1994), we p redicted that if high (unrealistic) optimism is

a risk factor among college students, AR would benefit these students in compar-

ison to their counterparts who did not receive A R. Specifically, overly optimistic

students who received AR would have less test anxiety, higher cum ulative grade

point averages, and withdraw from fewer courses than would overly optimistic

s tudents who did not receive AR . T hus, the opt imism t ra it i t se l f would not

change; rather, achievement-related causal attributions would be altered so that

the positive expectations of overly optimistic students would gain a more realistic

basis. Further, if high optimism is a risk factor, we also expected that the positive

impact o f AR would apply to highly optimistic students, but not to less optimistic

students. Specifically, high-optimism students who received A R would differ sig-nificantly in outcome m easures, com pared to those high-optimism students w ho

did not receive the intervention; whereas low-optimism students who received

AR w ould not di f fer s ignif icant ly in ou tcom e measures , compared to low-

optimism students who did not receive AR.

Method

Participants

A sample of 236 (156 female, 57 male, 23 gender not sp ecified) first-year stu-dents fi-om a large, m idwestern university w as recruited from several sections of

an Introductory Psychology course in exch ang e for experimental credit. Partici-

pants’ ages ranged from 17 to 42 years ( M d n = 19).

Measures

Dispositional optimism. At the beginning of the academ ic year, students com-

pleted seven items adapted fro m Sche ier and C arver’s (1985) Life Orientation

Test (LOT), a Likert-style m easure o f dispositional optim ism. The LO T is a well-established measure that has been used to assess the role of optimism in a wide

variety of domains, such as health, athletic, and acad em ic settings (Chang, 1998;

Hellandsig, 1998 ; Stoecker, 1999). Four of the items of the LOT are worded in a

positive way (e.g., “In uncertain times, I usually expect the best”), and three of



OPTIMISM AND AT RIBU TION AL RETRAINING 715

the items are worded in a negative way (e.g., “I hardly ever expect things to go

my way”), with responses ranging from 1 (strongly disagree) to 10 (strongly

agree; Cronbach’s internal reliability, a = .78; test-retest reliability after 13

weeks, r = .72; Scheier & Carver, 1985). Responses for the negative items were

reverse-coded, and then scores on each item were summed so that high scores

reflect high levels of optimism and low scores reflect low levels of optimism

( M = 36.59, SD = 9.62: low-optimism group, M = 28.65, S D = 6.26; high-

optimism group, M = 43.84, SD = 5.55.

Attributional retraining (AR). Students were assigned to either an AR condi-

tion ( n = 184) or no-AR control condition ( n = 52), based on the section of the

Introductory Psychology course from which they were recruited. The AR condi-

tion consisted of an informational session emphasizing the positive effects of

effort attributions on college performance, as opposed to ability attributions, and

that ability-related performance is unstable and would likely improve throughout

their educational development (Menec et al., 1994; Wilson & Linville, 1982).AR

was presented to students in one of three ways: via videotape; videotape followed

by a brief group discussion; or a handout. Participants in the videotape-onlyAR

condition ( n = 70) viewed a brief 8-min film about two students discussing their

academic failure experiences. One student explained to the other that after per-

forming poorly in his courses, he began to put more effort into studying and hisgrades improved accordingly. At the end of the video, the importance of increas-

ing effort to enhance performance is emphasized.

Students assigned to the video-and-discussion AR group ( n = 44) watched the

same film as the previous AR group, which was followed by a 20-min discus-

sion. The discussion consisted of the students talking about their own success and

failure experiences, while the experimenter explained the importance of using

adaptive attributions, such as effort and strategy, to explain their academic per-

formances. The third AR group (handout only; n = 5 6 ) was given a one-page

handout that summarized the benefits of changing dysfunctional causal attribu-tions for failure (ie. , lack of ability) to functional attributions (i.e., lack of effort)

and was instructed to read the handout carefully. The control group completed a

filler questionnaire that took approximately the same amount of time to complete

as the AR sessions.A similar AR design has been used successfully in previous

academic research (Perry & Penner, 1990).

Graduating high school percentage. At the beginning of the year, high school

achievement was assessed by obtaining the mean percentage of all course grades

from the students’ senior year of high school ( M = 75.81%, SD = 8.65). High

school percentage was calculated by averaging students’ final grades in their col-lege entrance courses (English, mathematics, chemistry, and physics), which

were obtained from institutional records. The overall high school percentage

ranged from 57% to 95%. Previous research has shown that self-reported

high school grades correlate strongly with students’ final course grades in college



716 RUTHIG ET AL.

( r = .54; e.g., Perry e t al., 2001). By using students’ actual high school averages,

rather than their estimates, we hoped to improve the accuracy of our estimates of

students’ aptitudes.

Test anxiety Toward the en d of the acad em ic year, the students completed a

37-item true/false measu re of test anxiety adap ted from Sarason’s (1 975 ) Test

Anxiety Scale, a widely used ind icator of test anxiety in academ ic settings (e.g.,

Hamm ermaster, 1989; Jain, 1986). Th e Test Anxiety Scale includes items such a s

“I wish exam s did not bother me so much” ( M = 53.54, SD = 7.03; a = .80).

Responses for each i tem were summed so that high scores reflect high test

anxiety and low er scores indicate lower test anxiety.

Cumulative grade point average (GPA). Academic achievement was defined

as students’ cumulative GPA at the end of the school year (e.g., 4.5 = A+, 4.0 =

A, 3.5 = B+; M = 2.5 I , SD = 1.06).GPA is calculated by sum ming and averaging

final grades for each o f the cou rses in which students were en rolled. Final grades,

in turn, are determined by av eraging test scores, assign m ent mark s, and final

exam scores within each course. Thus, GPA represents an aggregate of students’

academ ic achievements across all courses for the entire school year.

Voluntary course withdrawal (VW). Attrition was assessed by the num ber of

credit hours studen ts dropped during the year, where 3 credit hours were equiva-

lent to a one-sem ester course and6

credi t hours w ere equivalent toa

two-semester course ( M = 3.29, SD = 5.12). Generally, students are perm itted to

enroll in a minimum of one 3-credit-hour course ( 3 hours) per academ ic year, or a

max imum of five 6-credit-hour courses (30 hours) per academ ic year. Prior to

dropping any courses, students in the current study were enrolled in an average of

22.5 credit hours for the academ ic year.

Procedure

The current study was conducted in three phases. Phase I took place approxi-mately 1 month into the academ ic year, timed intentionally to ensure that all stu-

dents had written and received feedback on their first Introductory Psychology test.

All participants signed up for a session time to com plete the surveys, which w ere

administered to groups of 25 to 50 students per session. The questionnaire used in

the first phase of the experiment included a m easure o f dispositional optimism.

In Phase 2 (approximately 1 mo nth later), all participants were assigned to

either an A R condition or to a no-AR control condition. Depending on the condi-

tion to w hich they were assigned, participants attended an AR session or com-

pleted a filler questionnaire.In Phase 3, which took p lace toward the end of the academ ic year, all stu-

dents com pleted a questionnaire that included a measu re o f students’ test anxiety.

At the end of the acade mic year, part icipants’ cumulative GPA and VW were

obtained from institutional records.




Results

Rationale for Analyses

Because we were interested in examining unrealistic optimism as a potential

r isk factor, w e sorted students into two categories: low and high (unrealistic)

optimism groups. For all median splits, scores at or above the median w ere clas-

sified as high. Median splits were chosen as our analytic approach b ecause w e

were prim arily interested in examining the m ore dist inct categories of low or

high optimism, sinc e they are o f greatest interest to the research l i terature.3

Although a dropped median wou ld have been optimal for these analyses, it was

not possible in the current study because when the median was dropped, samplesizes in the low- and high-optimism groups were too small.

Optimism (Low vs. High) by AR (AR vs. No AR ) 2 x 2 factorial ANC OVA s

were used, wi th cum ulat ive GPA and VW as dependent m easures , and high

school percentage as the covariate. The effects of optimism and AR on students’

test anxiety were assessed using an Optimism (Low vs. H igh) x AR (AR vs. No

AR) 2 x 2 factorial ANOV A without high school percentage a s a covariate. High

school percentage w as not included a s a covariate for the test-anxiety analyses

for two reasons. From a statistical standpoint, because it w as uncorrelated with

test anxiety, including high school percentage as a covariate in the Optimism x

AR interaction would have decreased the power of the F stat ist ic. M oreover,

whi le high school achievemen t may be a go od indicator of future academic

achievement or course withdrawal (e.g., Perry et al., 2001), it does not necessar-

ily predict students’ test anxiety to the same extent. Because this study is in part

exploratory in nature, a liberal significance level @) of .10 wa s adopted for the

analyses due to our concern that a more conservative significance level would

result in comm itting Type I1 errors, thus obscuring important effects.

Preliminay Analyses

In order to en sure that differences in grading practices or teaching methods

between course instructors were not confound with the effects of optimism and

AR on the dependent measures, a preliminary one-way ANOVA was run with

course instructor as the independen t variable; and cumulative G PA, VW, and test

anxiety as the de pend ent measures . No s ignif icant di f ferences were found

3For those readers interested in a three-group optimism analytic model, preliminary analyses

revealed that splitting stude nts into three optimism groups (low, moderate, and high optimism) did

not add any new information to our findings. Specifically, all three groups were similar in academ ic

achievement ( M = 2.84, 2.44, and 2.62 for low, moderate, and high optimism, respectively), F ( 2 ,

42) = 0.84, ns; test anxiety ( M = 3.00,4.88, and 5.57, respectively), F(2 ,45) = 0.49, tis; and VW rates

( M = 54.44, 53.50, and 55.25, respectively), F( 2,2 9) = 0.12, ns.



718 RUTHIG ET AL.

between course instructors of different Introductory Psychology classes on any of

the dependent measures: cumulative GPA, F(5 , 197) = 0.45, ns; VW, F(5 ,211 ) =

0.87, ns; and test anxiety, F ( 5 , 123) = 1.60, ns, indicating that the effects of the

main analyses would not be due to instructor effects.

Attributional retraining (AR) conditions. Preliminary one-way ANOVA s

revealed no significant differences between the three types of AR condit ions

(video, video-and-discussion, handout) on an y of the dependent variables. Spe-

cifically, for GPA, the hando ut group , video group, and video -and-disc ussion

gro up did not differ significantly ( M = 2.55, 2.78, an d 2.83, respectively), F(2 ,

167) = 1.73, ns. Similarly, these groups did not differ on either VW ( M = 3.75,

2.85, and 2.35, respectively), F( 2, 181) = 1.54, ns; nor d id they differ on test

anxiety ( M = 55.5, 52.59, and 51.67, respectively), F( 2, 104) = 2.95, ns. Thus, all

three groups were com bined to create our single AR condition, which w as com-

pared to the no-AR (control) condition.

Correlations. The co rrelations between all variables are presented in Table 1.

Optimism was negatively related to test anxiety ( r = - .26, p < . O l ) , so that the

more optimistically students rated themselves, the less test anxiety they reported

throughout the academic year. In add ition to its relation to optimism, test anxiety

was correlated negatively with students’ academic achievement, as might be

expected: The m ore anxiety they exp erienced, the lower w as their cumulativeGPA (Y = -.29, p < .01). An exp ecte d positive correlation between high school

percen tage and cumulative GPA (1.= .61, p < . O O l ) and a negative correlation

with VW ( r = -.15, p < OS) indicated that higher achievemen t in high school was

associated wi th h igher col lege achievem ent and greater course pers istence.

Finally, students’ GPA and V W rates w ere negatively correlated, so that the bet-

ter their academic performance, the fewer courses they dropped ( r = -.20, p <

.Ol). It is important to note that the ab sence of correlations between high school

percentage w ith either AR or optimism validates the random assignment of stu-

dents into e i ther AR or control groups, and conf i rms that opt imism was notrelated to students’ academ ic aptitude (Table 1).

Ma in Analyses

Academic achievement. Table 2 presents the m eans and standard deviations

of the study variables. As indicated in Table 3, there was no m ain effect for opti-

mism on cumulative GPA, F ( 1 , 180) = 0.77, ns , but AR did have a significant

main effect, F( 1, 180)= 3.19, p = .08, qualified by an interaction with optimism,

F(1, 180) = 4.12, p < .05. Interestingly, in the control (n o AR ) condition, high-optimism students had a lower GPA ( M = 2.22) than did low-optimism students

( M = 2.72), t ( 36 ) = I .81, p = .08, indicating that high optimism may be a risk fac-

tor for students who do not receive A R. T his effect is notable because it was

detected despite the fact that it occurs with a global m easure of cumulative G PA,



Te1

Z

oOd

C

ao

Am

o

Su

V

a

e

Vae

1

2

3

4

5

6

1Omism

-

.0(2

.0(2-2(1

.0(2

-0(2

3 z

z

3Hgs

pca

- -1(1

.6(2

-1(2

D Z

2Aboreann

-

.0(2-1(1

.1(2

-1(2

(

4Taey

- -2(1

.1(1

O a

5

Gapnaa

-

-2(2

= D n5 m

-

6Vuayc

whaw

C

NoeVunpehreeonsz

9

*

< 05.*

< 0

r



720 RUTHIG ET AL.

2.8 -(u

0)

6 2.6->m-5 .4 -

p 2.2 -

n0)

(3

Table 2

Means a nd Standard Deviations of Study Variables

--.--ow

-m- High

Measures M SD

Hig h school percentage 75.81 8.65

Optimism 36.59 9.62

Cumulative grade point average 2.51 1.06

Voluntary co urse withdrawal 3.29 5.12

Test anxiety 53.54 7.03

I..,

NoARI

AR

Attributional retraining

Figure I . Effects of optimism and attributional retraining on students' cumulative gradepoint average.

a compilation of many tests and assig nm ents in numer ous courses, which poten-

tially contains a large amoun t of error. In fact, high -optim ism ho-A R students

had the lowest GPA o f the four groups (Figure 1). However , h igh-optimism

stude nts also benefited the most from the positive effects of AR : high-optimism

students who received AR performed significantly better ( M = 2.82) than did

high-optimism stud ents wh o did not receive AR ( M =2.22), t(94) = 2.56,p < .05

(Figure I ) . O f note is that this differen ce translates into mo re than a fetter grad edifference between highly optimistic students wh o received AR a nd those who

did not (e.g., a grade of B vs. C+). For low-optimism students, no differences in

GPA emerged between control ( M = 2.72) and AR conditions (A4 2.67), t (87)=

0.25, ns.



Te3

Mana

Inea

o

Ee

so

O

msma

A

rb

o

R

ran

n

(A

o

C

m

avGa

Pon

A

a

(G

V

u

a

C

Wihaw

(VW)a

T

A

ey

GPA

VW

Taey

Vae

d

MS

F

d

MS

F

d

MS

F

Hgs

pca

1

4716

1

8132

-

- -

AR

1

14

31

1

1665

1

1209

Omsm

1

03

07

1

6226

1

4103

AR xOmsm

1

18

41

1

1063

1

1141

0 3 2Z0

Noe

Nmaod 1foaF

es

T

.o*

< 0

*

< 0



722 RUTHIG ET AL.

1._2 30

2

--.--ow

+ igh

NoAR AR


Figure 2. Effects of optimism and attributional retraining on students’ voluntary course

withdrawal.

Voluntary course withdrawal (VW). Although optimism produced no signifi-

cant main effect, F( 1, 189) = 2.61, ns, AR had a significant main effect on volun-tary course withdrawal, F(1, 189) = 6.53, p < .05, where students w ho received

AR dropped fewer credit hours than did those who did not receive AR (Table 3).

The interaction between AR and optimism was also significant, F(1, 189) = 6.35,

p < .05. Multiple-comparison t tests reveal that for low-optimism students, AR

compared to no AR made no difference in num ber of credit hours dropped ( M =

3.26 vs. 3.24), t(92) = 0.02, ns, yet for high-optimism students, receiving AR

resulted in greater course persistence, when compared to high-optimism students

who did not receive AR ( M = 2.34 vs. 7.26), t(98) = 3.19,y < .O 1 (Figure 2 ) . The

difference between high- and low-optimism groups in the no-AR condition w asnot statistically significant ( M = 7.26 vs. 3.24), t(37) = 1.47,p = .15, however,

high-optimism studen ts in the no-AR g rou p dropped more than twice as many

credit hours as did low-optimism students in the control g r o u p a matter of prac-

tical significance to many college administrators. Th us, similar to achievem ent

findings, high optimism appears to be a potential risk variable for course attrition,

yet the risk is significantly decreased w hen high-optimism students are given AR .

Test [email protected] were no significant main effects for test anxiety, but again

the interaction between optimism and AR was significant (Table 3),F(1, 133) =

4.19, p < .05.Multiple-comparison t tests reveal that high-optimism students whoreceived AR experienced less test anxiety ( M = 51.6) than did high-optimism stu-

dents who did not receive AR ( M = 56.23), t(103)= 3 . 4 0 , ~ .01 (Figure 3). Fur-

ther, low-optimism students in the control group did not differ in reported test

anxiety from low -optim ism studen ts wh o received AR ( M = 53.96 vs. 56.22),



OPTIMISM AND AlTRIBUTIONAL RETRAINING 723

--.--ow

+ igh

0 ‘ I I

NoAR AR


Figure 3. Effects of optimism and attributional retraining on students’ test anxiety.

t (62)= 1.12, ns. In the control group, students with high optimism did not differ

on test anxiety from students with low optimism, t (30) = 0.58, ns. In this case,

unlike achievement and voluntary course withdrawal, optimism was not a risk

factor for test anxiety, but similar to achievement and VW, highly optimistic stu-

dents’ test anxiety benefited from the positive effects of AR.

Discussion

The current study makes an important contribution to the research on the pos-

itive benefits of AR (Menec et al., 1994; Perry & Penner, 1990; P e n y & Struth-

ers , 1994) by sho win g that , for over ly opt imis t ic s tudents , AR led to thei r

improved academic p erformance in a college setting. S ignificantly, the current

study extends our knowledge beyond previous AR research that used m ore lim-

ited m easures of academ ic achievement (final course grade) to include ou tcome

measures of cumu lative achievement (GPA) and voluntary course w ithdrawal

across all courses for that year. This is particularly impo rtant because academic

performance is not based on a single course outcome, but includes all of the stu-

dents’ courses for that year. Thus, the AR effect appears to ha ve been powerful

enough to positively influence students’ performance and retention, not only in

their Introductory Psychology course, but in all of their cou rses for that academ ic

year, as reflected in their cumulative GPA and V W rates. Further, this stud y

extends earlier research to include an important dispositional variable (optimism)

not com monly regarded as having academic risk qualities. Th erefore, this study

contr ibutes to previous research by suggest ing that AR can lead to higher

achievement, lower test anxiety, and greater persistence in college courses among

high-optimism students.



724 RUTHIG ET AL.

The current study also extends past research regarding the benefits and liabil-

ities of optimism. For instance, previous researchers have stated that positive illu-

sions may be beneficial in the short term, but no t in the long term (C olvin &

Block, 1994; Paulus, 1998). We addressed the long-term m erits o f optimism by

exam ining its effects on distal scholastic outcomes, such as students’ cum ulative

academic achiev em ent and voluntary cour se withdrawal ov er a full year in col-

lege. Robins and Beer (2001) also expressed a need for more external criteria to

measure the benefits of positive illusions. We addressed these concerns by again

using voluntary course withdrawal and cumulative academic achievement as out-

com e me asures, in addition to self-report mea sures of test anxiety. Thus, ou r

study demonstrates the diverse long-term consequences of optimism by compar-ing highly optimistic students and less optimistic students in term s o f their test

anxiety, VW, and cum ulative GPA over an entire academ ic year.

Accordingly, the possible benefits o f AR ar e quite notable in view o f the cur-

rent findings. Th at is, without A R, high-optimism students had the lowest cumu-

lative GPA and the high est VW rates of all four grou ps in the study. How ever,

simply giving high-optim ism studen ts a single AR session resulted in high-

optimism students achieving the highest cum ulative achievemen t (GPA), the

lowest V W rates, and the lowest level of test anxiety, compared to the other three

groups of students.

Optimism and Attributional Retraining

It is notable that high optimism appeared to be a risk factor for the college stu-

dents in our study. For instance, high-optimism students who did not receive A R

had the lowest cumulative GPA of all groups and the highest V W rates. Further,

the difference between high- and low-optimism studen ts who did not receive A R

translated into a whole letter grade difference (i.e., the difference between a C+

and a B) and twice a s many credit hours dropped! H owever, w hen high-optimismstudents were given A R, their GPA significantly increased, their VW signifi-

cantly decreased, and their test anxiety significantly decreased. These findings are

consistent with previous research that suggests that AR is most effective among

students wh o are at risk (Men ec et al., 1994; Perry & Struthers, 1994). Thus, in

the current study, high-optimism students were a t risk in terms of their academic

achievement and course withdrawal, but A R w as also the most effective amo ng

this risk group, compared to less optimistic students. These findings suggest that

teaching optim istic students that controllable factors such a s effort or strategy are

responsible for their academ ic achievemen t, rather than uncontrollable factorssuch as ability can h elp these stude nts to significantly improve their acade mic

performance, lower their test anxiety, and persist in their college courses.

Th e current findings give further insight into the o ngoing optimism debate.

Specifically, although we induced AR in a laboratory setting, its application was




designed to unfold in actual classroom settings. Once high-optimism students are

attributionally retrained and learn to make controllable, unstable causal attribu-

tions, they may then apply these adaptive causal attributions to future positive

and negative academic outcomes and even to outcomes in their daily lives. Thus,

these individuals may come to realize that controllable causes are responsible for

their success. However, without the positive effects of AR, these same individ-

uals may not have the necessary “grounding” to help prevent unrealistic expecta-

tions in their daily lives. Specifically, Higgins et al . (1997) suggest that

unrealistic optimism is mediated by perceived control. Therefore, AR can help

high-optimism individuals gain more control by inducing controllable attribu-

tions thus attenuating the potential harm of unrealistic optimism.

Recall that the current study involved first-year college students, whose

experiences with the realities of university life can be extremely limited. As

Perry et al. (2001) note, the transition to university is particularly fraught with

unaccustomed failure for a variety of reasons, including an inability to adjust to

the increased demand for self-initiative and autonomy. Therefore, it is possible

that optimism during this novel transition period would be a greater risk factor,

compared to optimism in situations where individuals have previous experience

on which to base their expectations for success. By altering the maladaptive

(uncontrollable) causal attributions of optimistic first-year college students to

adaptive (controllable) causal attributions via AR, the benefits of optimism may

be maximized and the costs greatly reduced, resulting in fewer subsequent failure

experiences.

Although the mean level of optimism among the high-optimism group in this

study was not exceptionally high on our scale (A4 43.84), nevertheless it

allowed us to differentiate the optimism variable and to demonstrate consistent

and reliable results. Therefore, we would expect that whatever negative patterns

occurred for our definition of “high” optimism would be more pronounced had

we been able to specify a high-optimism group having an even higher mean scoreon our scale (e.g., M = 60). Unfortunately, in order to retain sufficient statistical

power, given our sample size, we had to restrict our specification of “high” opti-

mism. It should be noted that our findings for such a “moderate” level of high

optimism are restricted to those groups having the unique features of our

transitional first-year students. It may be that the potential risks of being overly

optimistic are especially prevalent during certain transition periods in indi-

viduals’ lives. If this is the case, one would expect to find similar results among

high-optimism first-year graduate students, newly hired faculty members (Perry

et al., 1997), or other populations where individuals have no previous experienceon which to base their achievement expectations. Further research focusing on

optimism levels in different life transitions would be helpful in increasing our

understanding of when optimism is a potential risk factor and when it is bene-

ficial.



726 RUTHIG ET AL.

Thus, the current study provides additional insight into the optimism debate

in that whereas high optimism m ay b e a risk factor am ong college students, the

risk can be am eliorated through A R by givin g students the tools to ground their

unrealistic exp ectation s and chan ge their attribution s from uncontrollable (i.e.,

luck, ability) to controllable ones (i.e., effort) for which they are responsible. So,

instead o f thinking “I will be successful this year” and not believing that this pos-

itive outcome is within one’s own control, attributionally retrained students wh o

are high in optimism can change their thinking to “I will be successful because I

will study hard, take good lecture notes, and work hard throughout the academic

year.”

This s tudy a l so ex tends prev ious t es t anxie ty research (Bembenut ty &

McKeachie, 2000; Choi, 1998; Hembree, 1988) by linking it to optimism and

AR, thereby examining the possibility that test anxiety may partially account for

students’ course withdraw al. Tha t is, the optimism-AR interaction on voluntary

course withdrawal may be the result of a decrease in test anxiety reported by the

high-optimism students wh o received A R. Specifically, providing AR to high-

opt imism s tudents l ed to bo th decreased volunta ry course wi thdrawal and

decr eased test anxiety. Interestingly, a subs equ ent regre ssion analys is revea led

that, of the different academ ic variables (optimism, A R, test anxiety), test anxiety

was the best predictor of voluntary course withdraw al (p=

.25 ) , t(213)=

3.54,

Knowing that test anxiety significantly predicts voluntary course withdrawal,

it is possible that optimistic students who received AR persisted in their courses

because they also experienced a decrease in test anxiety. Recall that AR is

designed to change dysfunctional, uncontrollable causal beliefs (e.g., “I lack the

ability”) to functional, controllable causal beliefs (e.g., “1need to study more”) in

the presence of failure. Consequently, this shift in thinking m ay provide attribu-

tionally retrained students with both cog nitive strategies and a feeling of control

over their academ ic performan ce needed to alleviate their test anxiety. Subse-quently, these cognitive strategies and perceptions of control m ay he lp students

persist in their courses. In contrast, low-optimism students and students who did

not receive A R m ay have w ithdrawn from their courses because they did not

experience a sh ift in their causal beliefs or gain a sense o f academic control, and

thus their test anxiety was not reduced over the course o f the academ ic year.

p < .001.

Lirnilations

Although the results of the current study provide further insight into thebenefits of AR , they sh ould be interpreted with so me cau tion. Specifically, all

three types o f AR were comb ined to form the single AR condition in this study.

Thus, the videotape condition, videotape-and-discussion ondition, and handout

condi t ion w ere a l l summed together a s one AR con di t ion. Al though i t was



OPTIMISM AND A'TTRIBUTIONAL RETRAINING 727

appropriate to combine AR conditions in the current study, it is important to

note that past research has focused on find ing the m ost effective AR intervention

for different at-risk g rou ps (e.g., M ene c et al., 1994 ; Per ry & Struthers, 1994).

For exam ple, Hunter and Perry (1 996) found that a com bination of an AR video-

tape followed by an aptitude test was m ost effective for students with low h igh

school grades. Ha ll , Hladkyj, Perry, and R uthig ( in press) found that an AR

videotap e paired w ith a related written assignm ent wa s most effective a mo ng

low elaborative-learning students. Thus, past research on the effectiveness of

A R con ditions indicates that for different risk groups, certain types of AR are

more effective than are others. Consequen tly, further research is needed to deter-

m ine which of these specific condit ions is the m ost effective type o f AR for

students with high optimism or if , in fact , a new form of AR would be m ore

appropriate.

A second limitation of the curre nt study is that only low and high levels o f

optimism w ere addressed, rather than low, moderate, and high levels. The reason

for including only two groups in the current study was that the cell sizes became

too sm al l when spl i t in to three separate groups. Perhaps future s tudies can

provide a preliminary test o f optimism to ensure that there are sufficient numbers

of participants in each of the low, moderate, and high optimism g roups before

recrui ting them to par t ic ipate in an AR study. This addi tion of a m oderateoptimism group m ay pro vide insight into the optimal level of optimism where

students may have positive academic prospects without having unrealistic expec-

tations.

Keeping these limitations in mind, this study clearly illustrates the possible

benefi ts of AR amo ng first-year college students in terms of improving their

cumulative academic achievement, lowering their test anxiety, and reducing their

course withdrawal. Furthermore, the current findings also indicate that AR is par-

ticularly effective when combined with optimism. In this instance, being op timis-

tic is an asset because students already have positive ex pectations for the future,and AR provides the reality check by stressing that it is up to the student to make

those successful outcomes happen. Peterson (2000) summed up this idea of real-

istic optimism by suggesting that the benefits o f optimism exist, if approached in

an evenhanded w ay. Prov iding high-optimism students with AR appears to be an

effective means of app roaching optimism in this evenhanded w ay and thus trans-

forming unrealistic optimism into realistic optimism.

References

Aspinwall, L. G., Richter, L., & Hoffmann, R. R. (2000). Understanding how

optimism works. In E. C. Chang (Ed.), Optimism andpessimism: Implications

fo r theory, research, and practice (pp. 2 17-238). Washington, DC: American

Psychological Association.



728 RUTHIG ET AL.

Bembenutty, H., & McKeachie, W. J. (2000, A pril). Emotion regulation and test

anxiety: The contribution of academ ic delay of gratzjication. Paper presented

at the annual meeting of the American Educational Research Association,

New Orleans, Louisiana.

Carver, C . S., Reynolds, S. L., & Scheier, M. F. (1994). The possible selves of

optimists and pessimists. Journal of Research in Personaliq, 28, 133-141.

Carver, C. S.,& Scheier, M. F. (2000). Optimism , pessimism, and self-regulation.

In E. C. Chang (Ed.), Optimism and pessimism: Implications fo r theory,

research, andpractice (pp. 3 1-52). Washington, DC: A merican Psychological

Association.

Chang, E. C. (1998). Does dispositional optimism moderate the relation between

perceived stress and psychological well-being? A preliminary investigation.

Personality and Individual Differences, 25,233-240.

Domino, B., & Conway, D. W. (2000). Optimism and pessimism from a histori-

cal perspective. In E. C. Chang (Ed.), Optimism andpessimism: Implications

for theory, research, and practice (pp. 3-30). Washington, DC: American

Psychological A ssociation.

Choi, N. (1998). The effects of test format and locus of control on test anxiety.

Journal of College Student Development, 39 ,6 16-620.

Colvin, C. R., & Block, J. (1994). Do positive illusions foster mental health? Anexamination of the Taylor and Brown formulation. Psychological Bulletin,

Dougall, A. L., Hyman, K. B., Hayward, M. C., McFeeley, S., & Baum, A.

(2001). Optimism and traumatic stress: The importance of social support and

coping. Journal o f Applied Social Psychology, 31,223-245.

Floyd, C. (1997). Achieving despite the odds: A study of resilience among a

group of African American high school seniors. Journal of Negro Education,

Fromme, K., Katz, E. C., & Rivet, K. (1997). Outcome expectancies and risk-taking behavior. Cognitive Therapy and Research, 21 ,42 1-422.

Gillham, J. E., Shatte, A. J . , Reivich, K. J . , & Seligman, M. E. P. (2000). Opti-

mism, pessimism, and explanatory style. In E. c.Chang (Ed.), Optimism and

pessimism. Implications fo r theory, research, and practice (pp. 53-76).

Washington, DC: Am erican Psychological Association.

Hall, N. C., Hladkyj, S ., Perry, R. P., & Ruthig, J. C. (i n press). The role of attri-

butional retraining and elaborative learning in college students’ academic

development. Journal of Social Psychology.

Hammermaster, C. S. (1989). Levels of performance and cognitive interference intest-anxious subjects. Alberta Journal of Educational Research, 35, 164- 170.

Helgeson, V. S. (1999). Applicability of cognitive adaptation theory to predicting

adjustment to heart disease after coronary angioplasty. Health Psychologv,

116,3-20.

65, 181-188.

I H , 561-569.




Hellandsig, E. T. (1998). Motivational predictors of high performance and dis-

continuation in different types of sports among talented teenage athletes.

International Journal of Sport Psychology, 29,27-44.

Hembree, R. (1988). Correlates, causes, effects, and treatment of test anxiety.

Review of Educational Research, 58,47-77.

Higgins, N. C., St. Am and, M. D., & Poole, G. D. (1997). The controllability of

negative life experiences mediates unrealistic optimism. Social Indicators

Research, 42,299-323.

Hunter, A. J., & Perry, R. P. (1996, August). Attributional retraining: Identifiing

at-risk students before admission to university. Presented at the 26th Inter-

national Congress of Psychology, Montreal, Quebec, Canada.

Jain, S. (1986). An experimental study of anxiety and psychomotor performance.

Asian Journal of Psychology and Education, 17, 14-20.

Kao, G., & Tienda, M. (1995). Optimism and achievement: The educational per-

formance of immigrant youth. Social Science Quarterly, 76, 1-19.

Lai, J. C. L., & Wan, W. (1996). Dispositional optimism and coping with aca-

demic examinations. Perceptual and Motor S kills, 83,23-27.

Lennings, C. J. (2000). Optimism, satisfaction, and time perspective in the eld-

erly. International Journal of Aging and H uman Development, 51, 167- 181.

Menec,V.

H., Perry, R. P., Struthers, C. W., Schnwetter, D. J., Hechter, F. J.,&

Eichholz, B. L. (1994). Assisting at-risk college students with attributional

retraining and effective teaching. Journal of Applied Social Psychology, 24,

Paulus, D. L. (1998). Interpersonal and intrapsychic adaptiveness of trait self-

enhancement: A mixed blessing? Journal of Personality and Social Psychol-

Perry, R. P., Hechter, F. J., Menec, V. H., & Weinberg, L. (1993). Enhancing

achievement motivation and performance in college students: An attribu-

tional retraining perspective. Research in Higher Edu cation, 34, 687-720.Perry, R . P., Hladkyj, S., Pekrun, R., & Pelletier, S . T. (2001). Academic control

and action control in the achievement of college students: A longitudinal field

study o f self-regulation. Journal of Educational Psychology, 93, 776-789.

Perry, R. P., Menec, V. H., Struthers, C. W., Hechter, F. J., Schnwetter, D. J., &

Menges, R. J. (1997). Faculty in transition: A longitudinal analysis of the role

of perceived control and type of institutions in adjustment to postsecondary

institutions. Research in Higher Ed ucation, 38, 5 19-556.

Perry, R. P., & Penner, K. S. (1990). E nhancing academic achievem ent in college

students through attributional retraining and instruction. Journal of Educa-tional Psychology, 82, 262-27 1.

Perry, R . P., & Struthers, C . W. (1994, A pril). Attributional retraining in the col-

lege classroom: Some cause fo r optimism. Paper presented at the American

Educational Research Association annual m eeting, New O rleans, Louisiana.

675-701.

ogy, 74, 1197-1208.



730 RUTHIG ET AL.

Peterson, C. (2000). The future o f optimism. American Psychologist, 55,44-55.

Robins, R. W., & Beer, J. S. (2001). Positive illusions about the self Short-term

benefits and long-term costs. Journal of Personality and Social Psychology,

Sarason, 1. G. (1975). The Test Anxiety Scale: Concept and research. In I. G.

Sarason (Ed.), Stress and anxie@ (Vol. 5 , pp. 193-216). New York, N Y John

Wiley & Sons.

Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assess-

ment and implications of generalized outcome expectancies. Health Psychol-

Scheier, M.F.,

& Carver, C . S. (1987). D ispositional optimism an d physical well-

being: The influence of generalized outcome expectancies on health. Journal

o Personality, 55, 169-210.

Scheier, M. F., & Carver, C. S. (1993). On the power of positive thinking: The

benefits of being optimistic. Current Directions in Psychological Science, 2,

Stoecker, J. A. (1999). Optimism and grade expectancies. Psychological Reports,

Struthers, C. W., & Perry, R. P. (1996). Attributional style, attributional retrain-

ing, and inoculation against m otivational deficits. Social Psychology ofE d u-cation, I , 171-187.

Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social psycho-

logical perspective on mental health. Psychological Bu lletin, 103, 193-210.

Thompson, S . C., Sobolew-Shubin, A,, Galbraith, M. E., Schwankovsky, L., &

Cruzen, D. (1 993). Maintaining perceptions of control: Finding perceived

control in low control circumstances. Journal of Personality and Social Psy-

Weiner, B. (1986). An attributional theory o achievement motivation and emo-

Weiner, B. (1 995). Judgments of r e . ~ p o n s j ~ i l j ~ :foundation fo r a theory o

Weinstein, N. D. (1980). Unrealistic optimism abou t future life events. Journal of

Weinstein, N. D. ( 1 989). Optimistic biases about personal risks. Science, 246,

Wilson, T. D., & Linville, P. W. (1982). lmproving the academic performance of

college freshmen: Attributional therapy revisited. Journal of Personality and

Social Psychology, 42, 367-376.Wilson, T. D., & Linville, P. W. (1985). Improving the performance of college

freshmen with attributional techniques. Journal of Personality and Social

80, 340-352.

ogy, 4 , 2 19-247.

26-30.

84,873-879.

choloa, 64,293-304.

tion. New York, N Y Springer-Verlag.

social conduct. New York, N Y Guilford.

Personality and Social Psychology, 39, 806-820.

1232-1233.

P S y c h l o ~ ,9,287-293.

Optimism and Attributional Retraining

Documents

Transcript of Optimism and Attributional Retraining